Transform video frames using text instructions
Swap faces in images from a source to all targets
Edit images based on text instructions